Molecular multipole moments derived from collisional quenching of H(2s) Autor(en): Dose, V. / Semini, C. Objekttyp: Article Zeitschrift: Helvetica Physica Acta Band (Jahr): 47 (1974) Heft 6 PDF erstellt am: 12.05.2018 Persistenter Link: http://doi.org/10.5169/seals-114585 Nutzungsbedingungen Die ETH-Bibliothek ist Anbieterin der digitalisierten Zeitschriften. Sie besitzt keine Urheberrechte an den Inhalten der Zeitschriften. Die Rechte liegen in der Regel bei den Herausgebern. Die auf der Plattform e-periodica veröffentlichten Dokumente stehen für nicht-kommerzielle Zwecke in Lehre und Forschung sowie für die private Nutzung frei zur Verfügung. Einzelne Dateien oder Ausdrucke aus diesem Angebot können zusammen mit diesen Nutzungsbedingungen und den korrekten Herkunftsbezeichnungen weitergegeben werden. Das Veröffentlichen von Bildern in Print- und Online-Publikationen ist nur mit vorheriger Genehmigung der Rechteinhaber erlaubt. Die systematische Speicherung von Teilen des elektronischen Angebots auf anderen Servern bedarf ebenfalls des schriftlichen Einverständnisses der Rechteinhaber. Haftungsausschluss Alle Angaben erfolgen ohne Gewähr für Vollständigkeit oder Richtigkeit. Es wird keine Haftung übernommen für Schäden durch die Verwendung von Informationen aus diesem Online-Angebot oder durch das Fehlen von Informationen. Dies gilt auch für Inhalte Dritter, die über dieses Angebot zugänglich sind. Ein Dienst der ETH-Bibliothek ETH Zürich, Rämistrasse 101, 8092 Zürich, Schweiz, www.library.ethz.ch http://www.e-periodica.ch
Helvetica Physica Acta Vol. 47, 1974. Birkhäuser Verlag Basel Molecular Multipole Moments Derived from Collisional Quenching of H(2s) by V. Dose and C. Semini1) Physikalisches Institut der Universität Würzburg, Würzburg, Germany (10. VIL 74) Abstract. In this paper we try to improve on existing Born approximation treatments [1] of collisional quenching of métastable hydrogen atoms by molecules in thermal energy collisions. A four-state coupled-channel impact parameter calculation is carried out. The molecule is treated as a static charge distribution whose rotational motion is neglected. Cross-section results from the present calculation are used to calculate new values of multipole moments for CH3I, N2, C02, and CC14 from available experimental data. I. Introduction The well-known sensitivity of métastable atomic hydrogen to electric fields makes it a promising tool for the investigation of molecular charge distributions. This possi bility was first pointed out by Gersten [1] who calculated cross-sections for H(2s) + M ~> H(2p) + M -> H(ls) + Ly - a + M (1) in a first-order straight-line trajectory impact parameter approach. To calculate transitions for the above process Gersten replaced the molecule by its lowest-order non-vanishing multipole moment. This simplification is similar to the procedure adopted in calculating van der Waals forces, e.g. with overlap between charge distri butions of the interacting systems being neglected. Since the order of magnitude of quenching cross-sections is 10"14 cm2 corresponding to an interaction distance >10-7 cm, the van der Waals assumption should be well justified. However, due to the extreme sensitivity of métastable hydrogen to electric fields, a Born approximation treatment of the collision leads to transition probabilities exceeding unity already at these large impact parameters. Gersten, therefore, adopted the standard procedure from line broadening theory and took transition probabilities P(p) for impact parameters p < p0 equal to one, p0 being the largest root of the equation P(p) V (2) We consider this procedure to be formal and unphysical in the present case. From the interaction distance of 10-7 cm given above the interaction time for a projectile velocity of IO6 cm/sec would be of the order of IO-13 sec. Comparison of this time with») Now at Landis & Gyr AG, Zug, Switzerland.
624 V. Dose and C. Semini H. P. A. the life time of the n 2 states of atomic hydrogen in a strong electric field t «3.2 x IO-9 sec [2] shows that the effect of the collision will only be a thorough mixing of 2s and 2p states. The decay of the 2p state by emission of Lyman-Alpha radiation, which completes the quenching process, takes place when the collision is already over. As a consequence elastic collisions should occur even at impact parameters p < p0. This, in turn, would lead to smaller quenching cross-sections. In Section 2 below we give a short outline of the first-order theory modified by the strong mixing assumption. In Section 3 a coupled-channel solution following the method proposed by Takayanagi [3] and explored by Bauer and Callaway [4] is pre sented. Results are discussed in Section A. We shall use atomic units except where otherwise specified. II. First-Order Theory Consider a molecule M approaching a métastable hydrogen atom with velocity v (Fig. 1). Let R denote the position of the molecule's centre of mass with respect to the hydrogen atom. The angles specifying the direction of R are chosen such that p 0 is in the direction of?and 9 tt/2 fixes the collision plane. Let y and w specify the orient ation of the molecule with respect to the collision plane. The interaction matrix element VVpi Lp for hydrogen states with angular momentum quantum numbers L'p and Lp, respectively (U L ± 1) is then given by VVu.tLp ATTVT+lMt 2 (-1)«-'-1,/ / 1 /+l\ am \m a X J L'i Att \ YÎ+J9.P) 3 YÎ\Lp\Yr(X,w)--^TA>, (3) r is the radial coordinate in the hydrogen atom and / is the order of the molecule's first non-vanishing multipole moment Mt. In the special case of a 2so -* 2pp transition this co 0 v MOLECULE M H-ATOM Figure 1 Coordinates used to describe the collision.
Vol. 47, 1974 expression simplifies to Molecular Multipole Moments 625 / 1 / +1 \ Y,+r(0, p) V2so,2,» l2n(-iy+l-1vïtïml-2y!r(x,w)[ "J (A) \m p p mj Rl+2 The first-order transition amplitude is then * 2so,2p«J '2so,2pp,dt (5) We evaluate T2so2m in a straight-line trajectory approximation. With R p + vt (6) and coordinates chosen as in Figure 1 the time integration is given by AYi+n-r.9 R~2 dt: 0 for 1 + p m even l\ (-1) (Z + l)/2+ -M 21 + 3 vp'+1 \tt(1 + m+l-p)\(l + p+l-m) 1/2 (7) We insert (7) and (4) in (5) and obtain for the first-order transition probability Mt V 2so,2pu l+l \vp (12P)2-tt(1+1)(21 + 3)- i i i + i X' (_]\U+l)/2+X-m m yr(v.o>) m p p mj V(l + m + 1 p)\(l + p+1 m)\ (8) The sum in this expression is restricted to values of m such that I + p m odd. The transition probability, of course, diverges for small impact parameters p and is replaced by a constant for p smaller than some critical impact parameter pk. We define pkoy 2so,2pnOk)- 1 + K (9) and obtain the total cross-section as a function of (y, w). 1 f Q2so,2pu(x, <") - r jz rrpi + 2-TT J P2so2pil(p)pdp. (10)
626 V. Dose and C. Semini H. P. A. Finally, we sum over the degeneracy of the 2p state and average over molecular orientations Q.2S,2p= \ ^Q2so,2pAXi JI)dQ (H) and obtain Y2s,2p ttci.1 + (i + A)'/<'+1>\ ; i\(9-22l+3(i + i)(i\y) l/ci+i) (21 + 2)\ I V 2/(1+1) (12) Values of the constants C/ which arises from the numerical integration involved in (11) are given in Table I. Eq. (12) may be modified to obtain Gersten's result taking K 0 andc; 1. The choice K 0 corresponds to complete quenching for impact parameters p < po- For reasons given in Section I we propose to assume complete mixing of the n 2 states inside a critical impact parameter, e.g. K=l. This choice reduces the quenching cross-sections by a factor V2 for a dipole molecule and by a factor of 2 for I-+00 as compared to Gersten's result. A constant C, 1 is obtained in Gersten's work because he takes the average over molecular orientations and the sum over the degeneracy of the 2p state after equation (8). This is certainly incorrect though numerically of minor importance. III. Unitary Calculation The method proposed by Takayanagi [3] as described by Callaway and Bauer [4] consists of taking the matrix e~it (13) with T as defined in (5) as an approximation to the S-matrix. If we take into account only the n 2 states in hydrogen, T has the form 2s 2p-i 2^o 2Pi 2s 0 a b c 2p-i a* 0 0 0 2p0 b* 0 0 0 2p, c* 0 0 0 (14) The initial state 2s ~v 2p_x 0 îpo 0 AP» ^ 0 (15)
Vol. 47, 1974 Molecular Multipole Moments 627 is transformed to the final state 2s * cose 2^_i ia+/e sin E 2^o ib+/e sine 2px ic+/e sin E (16) with 0 2+ 6 2+ c 2}l/2. (17) From (13) we obtain the transition probability P2s->2p sln2e (18) and 72 _ A I 2so2puI U 1 (19) It is worthwhile to recall the Born approximation result (8) which is in the present notation 2s->2p(Born) E (20) With (4) we obtain explicitly 12 Mt aixia AAy y Yr(x,co) V(l + m)\(l-m)\ 2-11/2 (21) The prime on the inner summation indicates that m is restricted to values such that 1 + p mis odd. Using (21) in (18), the total cross-section is obtained by an integration over all impact parameters. The remaining average over molecular orientations must be carried out numerically. The result is <?, *r i, TT \ 1 TT \1/(,+1) Y/('.+1) (M,)2K'+1) (MA cos (22) with s^=0 y' YT<X.<») V(I + m)\ (I - m)\ 2 \ 1/(1+1) sin x dx dw. (23) Values of Bt are given in Table I.
628 V. Dose and C. Semini H. P. A. IV. Results and Discussion The cross-sections obtained in Sections II and III may be conveniently written as (M, \2/0+1) Qi=Yi{-y\ (24) Values of y, are given in Table I for formula (22), (12) with K 0, and (12) with K =1. Comparison of the last two lines in this table shows that the simple 'strong mixing' estimate of Section II produces results which differ by less than 10% from the much more elaborate coupled-channel calculation. Equation (24), together with Table I, may be Table I Numerical constants occurring in formulae (12), (22), and (24) of the text. With Af, in esu and v in cm/sec /1 the cross section is given in cm2 2 3 4 Ref. c, 0.961 0.913 0.874 0.847 (12) B, 4.822 4.304 3.186 2.333 (22) Yi 30.8 12.6 8.44 6.72 (12, K 0) Yi 21.7 7.9 5.02 3.83 (12,K=1) Yi 23.3 8.15 4.90 3.61 (22) used to calculate multipole moments M, from experimentally determined quenching cross-sections. Table II lists cross-sections for an impact velocity of IO6 cm/sec from recent measurements of Dose and Hett [5]. The coupled-channel calculation (Table I, last line) is used to derive multipole moments M,. Currently recommended values [6] are listed for comparison. Table II Absolute quenching cross-sections for an impact velocity v 106 cm/sec in units of 10"14 cm2. The respective multipole moments must be multiplied by FACTOR to obtain their values in cgs units CH3I C02 N2 CC14 1 1 2 2 3 0i 4.7 2.0 1.2 0.83 M, 0.84 5.0 2.3 12 M,, rec 1.6 4.3 1.5 10 FACTOR io-18 10-26 io-2«io"3 There is a considerable disagreement between present and accepted values for the dipole moment of CH3I. Though the difficulties associated with pressure measurements of condensable vapours might introduce an appreciable error in the measured crosssection, it is hard to believe that this would account for a factor of two. The case of C02 is much more satisfactory. Noting that published values for the quadrupole moment of this molecule range from 1.7 [7] to 5.9 [8] the agreement, in fact, is good.
Vol. 47, 1974 Molecular Multipole Moments 629 A huge amount of data is available for the quadrupole moment of N2. But again the scatter of values is large, ranging from 0.8 [7] to 3.1 [9]. In the absence of a detailed discussion of the merits of the various methods to determine molecular multipole mom ents the present value of 2.3 must be considered a reasonable answer. Only three measurements of the octupole moment of CC14 are known at present. Values of 0.55 [10], 1.5 [11], and 3.0 [12] are compatible with our result of 1.2. The failure of the present method in the determination of the dipole moment of CH3I precludes any definite conclusions concerning its usefulness. Instead we recall that several more or less serious simplifications have been made. 1) Neglect of molecular rotational motion. 2) Neglect of higher order multipole moments. 3) Complete neglect of dispersion forces. While we are going to remove the first two restrictions with the aim to improve on the present treatment, a proper treatment of dispersion forces would at present meet with unsurmountable difficulties. Acknowledgement One of us (CS.) wants to thank the 'Schweizerischer Nationalfonds' for a research grant. REFERENCES [1] J. I. Gersten, J. Chem. Phys. 51, 637 (1969). [2] H. A. Bethe and E. E. Salpeter, Quantum Mechanics of One- and Two-Electron Atoms, (Academic Press, New York 1957). [3] K. Takayanagi, Progr. Theoret. Phys. (Kyoto), Suppl. 25, 1 (1963). [4] J. Callaway and E. Bauer, Phys. Rev. 140, A1072 (1965). [5] V. Dose and W. Hett, Proc. of the Fourth International Symposium on Molecular Beams, Cannes, 1973. [6] D. E. Stogryn and A. P. Stogryn, Molecular Phys. 11, 371 (1966); and private communica tions with Dr. D. E. Stogryn. [7] H. Feeney, W. Madigosky and B. Winters, J. Chem. Phys. 27, 898 (1957). [8] G. Birnbaum and A. A. Maryott, J. Chem. Phys. 36, 2032 (1962). [9] J. S. Murphy and J. E. Boggs, J. Chem. Phys. 49, 3627 (1968). [10] S. K. Garg, J. E. Bertie, H. Kilp and C. P. Smyth, J. Chem. Phys. 49, 2551 (1968). [11] S. Külich, Phys. Lett. 27A, 307 (1968). [12] D. L. Weinberg, J. Chem. Phys. 47, 1307 (1969).